alandtse / alexa_media_player Goto Github PK

View Code? Open in Web Editor NEW

1.4K 1.4K 274.0 2.43 MB

This is a custom component to allow control of Amazon Alexa devices in Home Assistant using the unofficial Alexa API.

License: Apache License 2.0

Python 99.53% Dockerfile 0.47%

alexa hacktoberfest home-assistant homeassistant

alexa_media_player's People

Contributors

Stargazers

Watchers

Forkers

snipercaine edstace crowland88 eugenelai chali272 kwdavidson aneisch niemyjski minimad-diver jyjx85 martokk goliath2k daniel-matthews stanosaka angel12 skysurf airdrummingfool dieterclaeys filhgd clach04 madsbekker l-lab sylph meatheadmike binslum3 refael65 tbruck martynrees c3reeman remcotim vageesh79 mviamin brianhanifin itzalgd24 thingsjar tomh05 rbhr liorofir slamdf150xc yourcoachzach herveb arkadije lindeluci omerdagan kbsandeep g3ppyt vswraith dungdt2016 akinomeroglu vatob thefuzz4 ricksmith2121 matbrewer396 fleshinachair robbiemacgit grinco erosennin82 playedin guennikasse mike6460 madpsy caphm cheynespc ladysuee esabea choss ahmadzeena tomhoover sslopes machinae gadgetchnnel jonydebu solventarchitect gerardmccafferty terryhonn macbury xatr0z arsvis aouicher javiges ctbartolillo slomanl1 sdig ashdwells fjkhack alkaraz32 furetto72 jasongetsit hernanigarza lanealarson hellofromearth mls228 joselu81 softhawk2000 axel7778 beertje135 jhonnymnemonic320 areburg domjelha mikeage

alexa_media_player's Issues

Add component to custom_updater

As I mentioned before in the home assistant commnuity support thread to this component, it would be nice so have a possibility to autoupdate this. This would be possible if add this to the custom updater component.

https://github.com/custom-components/custom_updater?files=1

Refactor codebase and verify flow

This issue is to refactor the codebase and ensure that the authentication and component access flow throughout the component is consistent. Major areas are how the configurator is being handled.

Captcha loop for non-EN domains

Moving this issue to its own so we can declutter #14.

I haven't tested it, but perhaps we change the Accept-Language header to "*" so we can accept whatever Amazon gives us and avoid having to check per region.
@dreimer1986, can you check?

Allow for multiple accounts using the component

I have a use case where I have echos that are on distinct accounts to intentionally restrict them from smarthome features, so that my kids can't torture each other, but I would love to be able remotely control them. Can the ability to have multiple accounts be added?

Feature Request: Notification component

I'm rusty with python and have just started using home assistant a couple days ago, but I needed a notification component so I made this hackish custom component.

I'll be setting up the rest of my systems before I might come back to clean this up. Posting this just in case someone need something similar.
It might be easier for someone w/ more experience to clean it up.
There are some unnecessary imports, should probably confirm that the echo devices exist in hass, renamed to echo tts, have media_player.alexa dependencies set, and allow setting volume in config file too?

configurations:

notify:
  - id: alexa_tts
    name: Alexa TTS
    platform: alexa_tts
    alexa_devices:
      - echo_dot

notify/alexa_tts.py

"""
Support for the Alexa media player Alex as a notification service.
"""
import logging
import json

from homeassistant.components import media_player
from homeassistant.components.notify import (
    BaseNotificationService, ATTR_TITLE)
from homeassistant.const import (
    CONF_API_KEY, CONF_DOMAIN, CONF_RECIPIENT, CONF_SENDER)

_LOGGER = logging.getLogger(__name__)

DEPENDENCIES = ['media_player','logger']

ALEXA_DEVICES = 'alexa_devices'
SERVICE_NAME = "Alexa TTS Notification"
ERROR_PREFIX = SERVICE_NAME + " Service Error"

def get_service(hass, config, discovery_info=None):
    """Get the Alexa TTS notification service."""
    alexa_devices = config[ALEXA_DEVICES]

    alexa_service = AlexaNotificationService(alexa_devices, hass)
    if (alexa_service.devices_are_valid):
      return alexa_service
    else:
      _LOGGER.error(ERROR_PREFIX+"alexa tts notification service devices are invalid")
      
    return alexa_service

class AlexaNotificationService(BaseNotificationService):
    """Implement a notification service for the Alexa TTS notifier service."""

    def __init__(self, alexa_devices, hass):
      """Initialize the service."""
      _LOGGER.info(SERVICE_NAME + " is initializing")
      self.alexa_devices = alexa_devices
      self.hass = hass

    def devices_are_valid(self):
      _LOGGER.info(SERVICE_NAME + " is checking if devices are valid")
      """Check whether device exist."""
      if not self.alexa_devices:
        _LOGGER.error(ERROR_PREFIX + "alexa_devices parameter is empty or missing.")
        return False

      devices_are_valid = True
      for device in self.alexa_devices:
        if not device:
          _LOGGER.error(ERROR_PREFIX+"An empty device is passed.")
          devices_are_valid = False
          continue
        if (device not in hass.media_player):
          _LOGGER.error(("{}: Device {} is invalid.").format(ERROR_PREFIX, 
            str(device)))
          devices_are_valid = False
      return devices_are_valid

    def send_message(self, message="", **kwargs):
      title = kwargs.get(ATTR_TITLE)      
      data = '. '.join(filter(None, [title, message]))
      
      _LOGGER.info(SERVICE_NAME + " is sending message to alexa devices with data: " + data)
      
      for alexa_device in self.alexa_devices:
        _LOGGER.info(SERVICE_NAME + " TTS Service call on :" + alexa_device)
        self.hass.services.call('media_player', 'alexa_tts',
          {'entity_id':"media_player." + alexa_device,'message':data})

last_called is too slow?

I'm trying to make of the last_called attribute. Ie things like "Alexa Turn On Lights" / "Alexa Turn On TV" and then based on the echo that was used turn on the lights for that room.

This #65 suggest that @brianhanifin got it working. But in my test there a good 10 seconds delay until the correct echo is reported back correctly.

Is there any way to force a refresh of the last_called attribute?

Cheers

TTS not work

I cant get alexa tts in my echo plus. I want spanish tts .Im from mexico and I try to conect to http://alexa.amazon.com.mx but not work. It only work with amazon.com. But i get this error in log Updating alexa media_player took longer than the scheduled update interval 0:00:10 and sometimes this error Error executing service <ServiceCall tts.google_say AttributeError: 'NoneType' object has no attribute 'group' any idea?

no longer showing devices

can anyone help please getting this error

2018-12-24 12:40:15 ERROR (MainThread) [homeassistant.components.media_player] Error while setting up platform alexa
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/site-packages/homeassistant/helpers/entity_platform.py", line 128, in _async_setup_platform
    SLOW_SETUP_MAX_WAIT, loop=hass.loop)
  File "/usr/local/lib/python3.6/asyncio/tasks.py", line 358, in wait_for
    return fut.result()
  File "/usr/local/lib/python3.6/concurrent/futures/thread.py", line 56, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/config/custom_components/media_player/alexa.py", line 146, in setup_platform
    config.get(CONF_DEBUG))
  File "/config/custom_components/media_player/alexa.py", line 618, in __init__
    self.login_with_cookie()
  File "/config/custom_components/media_player/alexa.py", line 641, in login_with_cookie
    self.login(cookies=cookies)
  File "/config/custom_components/media_player/alexa.py", line 776, in login
    self._data = self.get_inputs(soup)
  File "/config/custom_components/media_player/alexa.py", line 654, in get_inputs
    for field in form.find_all('input'):
AttributeError: 'NoneType' object has no attribute 'find_all'

recently changed my password as i had an account compromised and now this won't work i have changed the password in my !secrets file

Lxml does not build on hassio

Lxml is not building in hassio so the dependency cannot be satisfied.

HA 0.87 broke my alexa

sense updated my homeassitant to 0.87 version cant set mediaplayer alexa

2019-02-07 22:55:38 DEBUG (SyncWorker_3) [custom_components.media_player.alexa] Trying cookie from file /config/alexa_media.pickle
2019-02-07 22:55:38 DEBUG (SyncWorker_3) [custom_components.media_player.alexa] Error loading pickled cookie from /config/alexa_media.pickle: An exception of type FileNotFoundError occurred. Arguments:
(2, 'No such file or directory')
2019-02-07 22:55:38 DEBUG (SyncWorker_3) [custom_components.media_player.alexa] No valid cookies for log in; using credentials
2019-02-07 22:55:40 DEBUG (SyncWorker_3) [custom_components.media_player.alexa] Get to https://alexa.amazon.com/api/devices-v2/device was redirected to https://www.amazon.com/ap/signin?showRmrMe=1&openid.return_to=https%3A%2F%2Falexa.amazon.com%2Fapi%2Fdevices-v2%2Fdevice&openid.identity=http%3A%2F%2Fspecs.openid.net%2Fauth%2F2.0%2Fidentifier_select&openid.assoc_handle=amzn_dp_project_dee&openid.mode=checkid_setup&openid.claimed_id=http%3A%2F%2Fspecs.openid.net%2Fauth%2F2.0%2Fidentifier_
2019-02-07 22:55:40 DEBUG (SyncWorker_3) [custom_components.media_player.alexa] Preparing post to https://www.amazon.com/ap/signin/143-6557016-2207107 Captcha: None SecurityCode: None Claimsoption: None VerificationCode: None
2019-02-07 22:55:41 DEBUG (SyncWorker_3) [custom_components.media_player.alexa] Error message: Important Message!
            To better protect your account, please re-enter your password and then enter the characters as they are shown in the image below.

i have also deleted alexa_media.pickle file to see if help and nothing

Cookie login failure does not attempt to log in with credentials

A bad saved cookie won't login as the if-else login test logic is improperly nested. There's no else logic to go to the main login logic. I'll fix it soon.

GPL-3.0 intake license not compatible with home-assistant Apache-2.0

If you're planning to upstream, you'll need to change your LICENSE to Apache-2.0 so that you can submit your addon later to the main tree. The issue is anyone who offers you a pull request will assume you're taking their code as GPL-3.0, which cannot be converted by you to Apache-2.0 later without explicitly asking for permission to convert. Asking to change licenses later is a pain particularly since you don't have a CLA in place. VLC Media player went through this and you can see the pain in that experience here. Based on the PR history, you seem clean of 3rd party code for now, so you may want to nip it in the bud.

Good job btw. Hope this succeeds.

amazon.it tested not full working

Imported beta echo dot in italian using "amazon.it" in configuration.yaml , echo is imported and show music, but don't accept tts commands. Fix it in next version if you can, thanks.

Handle verification request

For users without 2FA, Amazon will occasionally request a sms or email verification. This was the cause of at least one Captcha loop report.

alexa tts

Hello,
I had everything working fine, then i started getting errors so i updated the alexa.py file.. but now i home assistant can't find any devices and i get this error in the log:

"2018-07-18 17:04:08 WARNING (MainThread) [homeassistant.core] Unable to find service media_player/alexa_tts"

am i doing something wrong?

Best regards,
Diogo Ramos

[Feature Request] Support of custom updater

It would be great I could could support the custom updater.

Invalid syntax error

File "/config/custom_components/media_player/alexa.py", line 677
    xcept Exception as ex:
                  ^
SyntaxError: invalid syntax

Convert into a HA platform

Per comments in #60 (which could be a sensor) and #56 (which is odd for HA), we should start moving the component into a platform that will spawn all the components.

#51 also would be a good addition.

update the new paths?

its seems now the custom componets have diferent path, maybe update the structure?
home-assistant/core#20807
media_player/alexa.py to alexa/media_player.py

Fix selection of locale to be based on user

Currently, we hardcode EN-US as the locale for much of the selection. It appears to cause no issues so far, but it may in the future.

Spawned from #74 where umlauts were not processed correctly by Amazon.

input text on media_player.alexa_tts

would it be possible to add to the media_player.alexa_tts option to have a input text for alexa to speack the input text directly on the media player?

Feature Request: Last Alexa Called

Today I learned that alexa_remote_control.sh can retrieve the last alexa device that was called. So, I downloaded the code and looked at the last_alexa() function. After some experimenting I figured out that function generates the Alexa API URL of https://alexa.amazon.com/api/activities?startTime=&size=1&offset=1.

A sample of JSON response is provided below. But the path we want is activities.sourceDeviceIds.serialNumber! With the device's serial number we can direct TTS responses to the correct Echo. This is important to users like myself that have an Echo device in every major room of the house (Kitchen, Living Room, Family Room, Bed Room, etc.)

What do you say? Do you want to make an amazing upgrade to our favorite TTS component? :)

{
	"activities": [{
		"_disambiguationId": null,
		"activityStatus": "SUCCESS",
		"creationTimestamp": 1547351733581,
		"description": "{\"summary\":\"turn on the living room computer\",\"firstUtteranceId\":\"A7WXQPH584YP:1.0/2019/01/13/03/G090QU0674750NLT/55:31::TNIH_2V.bcbda961-b9c8-4243-9d2d-b1f2279d6f1bZXV/1\",\"firstStreamId\":\"A7WXQPH584YP:1.0/2019/01/13/03/G090QU0674750NLT/55:31::TNIH_2V.bcbda961-b9c8-4243-9d2d-b1f2279d6f1bZXV\"}",
		"domainAttributes": null,
		"domainType": null,
		"feedbackAttributes": null,
		"id": "A12BCDEF3GH4I5#1547351733581#A7WXQPH584YP#G090QU0674750NLT",
		"intentType": null,
		"providerInfoDescription": null,
		"registeredCustomerId": "A12BCDEF3GH4I5",
		"sourceActiveUsers": null,
		"sourceDeviceIds": [{
			"deviceAccountId": null,
			"deviceType": "A7WXQPH584YP",
			"serialNumber": "G090QU0674750NLT"
		}],
		"utteranceId": "A7WXQPH584YP:1.0/2019/01/13/03/G090QU0674750NLT/55:31::TNIH_2V.bcbda961-b9c8-4243-9d2d-b1f2279d6f1bZXV",
		"version": 1
	}],
	"endDate": 1547351733581,
	"startDate": 1547351733581
}

`last_alexa()` from alexa_remote_control.sh

last_alexa()
{
${CURL} ${OPTS} -s -b ${COOKIE} -A "Mozilla/5.0" -H "DNT: 1" -H "Connection: keep-alive" -L\
 -H "Content-Type: application/json; charset=UTF-8" -H "Referer: https://alexa.${AMAZON}/spa/index.html" -H "Origin: https://alexa.${AMAZON}"\
 -H "csrf: $(awk "\$0 ~/.${AMAZON}.*csrf[ \\s\\t]+/ {print \$7}" ${COOKIE})" -X GET \
 "https://${ALEXA}/api/activities?startTime=&size=1&offset=1" | jq -r '.activities[0].sourceDeviceIds[0].serialNumber' | xargs -i jq -r --arg device {} '.devices[] | select( .serialNumber == $device) | .accountName' ${DEVLIST}
}

Use config flow instead of configurator for configuration

The use of the configurator is very hackish right now. It appears there is a push for components to move to config flow which appears in the integrations panel.

I haven't found the definitive dev guide on config flow yet, but it's mentioned in the integration score guide.

Here's one component that uses it.

If anyone sees the right document or thread, please post it here.

Switching account in configuration does not change logged in account

Changing the configuration settings will continue to use prior cookies to log in. We should perform a check on logged in user and if it doesn't match, log out and delete the cookie. We recently added a check for the logged in user so that can be reused.

This probably makes sense to deal with the multi-account login logic #37 at the same time.

ImportError with home-assistant v. >= 0.88.0

Error loading custom_components.media_player.alexa. Make sure all dependencies are installed
Traceback (most recent call last):
  File "/usr/local/lib/python3.7/site-packages/homeassistant/loader.py", line 147, in _load_file
    module = importlib.import_module(path)
  File "/usr/local/lib/python3.7/importlib/__init__.py", line 127, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 1006, in _gcd_import
  File "<frozen importlib._bootstrap>", line 983, in _find_and_load
  File "<frozen importlib._bootstrap>", line 967, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 677, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 728, in exec_module
  File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed
  File "/config/custom_components/media_player/alexa.py", line 16, in <module>
    from homeassistant.components.media_player import (
ImportError: cannot import name 'MEDIA_TYPE_MUSIC' from 'homeassistant.components.media_player' (/usr/local/lib/python3.7/site-packages/homeassistant/components/media_player/__init__.py)

Report

Caused by change in package bundling:

Note for custom component developers: We are moving to a new file structure. Platforms now live embedded in components. Custom platforms will have to be updated to follow this pattern. This is a breaking change in case your custom platform overrides a built-in platform. Rename your custom platform from, ie light/hue.py to hue/light.py.

alexa.py throws exception after upgrade to HA 0.78.0

First I want to thank you for your hard work making alexa tts work in HA.

Just now I upgraded HA to version 0.78.0 and Alexa media_player is gone from dashboard I am now having exception in the logs:

2018-09-18 19:09:09 ERROR (MainThread) [homeassistant.components.media_player] Error while setting up platform alexa
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/site-packages/homeassistant/helpers/entity_platform.py", line 128, in _async_setup_platform
    SLOW_SETUP_MAX_WAIT, loop=hass.loop)
  File "/usr/local/lib/python3.6/asyncio/tasks.py", line 358, in wait_for
    return fut.result()
  File "/usr/local/lib/python3.6/concurrent/futures/thread.py", line 56, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/config/custom_components/media_player/alexa.py", line 142, in setup_platform
    config.get(CONF_DEBUG))
  File "/config/custom_components/media_player/alexa.py", line 607, in __init__
    self.login_with_cookie()
  File "/config/custom_components/media_player/alexa.py", line 630, in login_with_cookie
    self.login(cookies=cookies)
  File "/config/custom_components/media_player/alexa.py", line 701, in login
    if (cookies is not None and self.test_loggedin(cookies)):
  File "/config/custom_components/media_player/alexa.py", line 673, in test_loggedin
    '/api/devices-v2/device')
  File "/usr/local/lib/python3.6/site-packages/requests/sessions.py", line 525, in get
    return self.request('GET', url, **kwargs)
  File "/usr/local/lib/python3.6/site-packages/requests/sessions.py", line 512, in request
    resp = self.send(prep, **send_kwargs)
  File "/usr/local/lib/python3.6/site-packages/requests/sessions.py", line 644, in send
    history = [resp for resp in gen] if allow_redirects else []
  File "/usr/local/lib/python3.6/site-packages/requests/sessions.py", line 644, in <listcomp>
    history = [resp for resp in gen] if allow_redirects else []
  File "/usr/local/lib/python3.6/site-packages/requests/sessions.py", line 192, in resolve_redirects
    proxies = self.rebuild_proxies(prepared_request, proxies)
  File "/usr/local/lib/python3.6/site-packages/requests/sessions.py", line 276, in rebuild_proxies
    environ_proxies = get_environ_proxies(url, no_proxy=no_proxy)
  File "/usr/local/lib/python3.6/site-packages/requests/utils.py", line 763, in get_environ_proxies
    return getproxies()
  File "/usr/local/lib/python3.6/urllib/request.py", line 2480, in getproxies_environment
    for name, value in os.environ.items():
  File "/usr/local/lib/python3.6/_collections_abc.py", line 744, in __iter__
    yield (key, self._mapping[key])
  File "/usr/local/lib/python3.6/os.py", line 669, in __getitem__
    raise KeyError(key) from None
KeyError: 'OPENBLAS_MAIN_FREE'

Could it be caused by breaking changes in Home Assistant or is it just unsuccessful upgrade that would require for me to clear something and authenticate again?

Updating alexa media_player took longer than the scheduled update interval

Need to figure out how to avoid this error. I don't think it's a functional problem and we can still get accepted into HA with this error but it's not desirable.

TypeError: argument of type 'NoneType' is not iterable

Running 0.9.0. Error is added to the log once per minute.

Error doing job: Future exception was never retrieved
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/concurrent/futures/thread.py", line 56, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/config/custom_components/media_player/alexa.py", line 167, in <lambda>
    track_utc_time_change(hass, lambda now: update_devices(), second=30)
  File "/usr/local/lib/python3.6/site-packages/homeassistant/util/__init__.py", line 325, in wrapper
    result = method(*args, **kwargs)
  File "/usr/local/lib/python3.6/site-packages/homeassistant/util/__init__.py", line 325, in wrapper
    result = method(*args, **kwargs)
  File "/config/custom_components/media_player/alexa.py", line 189, in update_devices
    update_devices, url)
  File "/config/custom_components/media_player/alexa.py", line 253, in __init__
    self.refresh(device)
  File "/config/custom_components/media_player/alexa.py", line 293, in refresh
    self._session['progress'] else None)
TypeError: argument of type 'NoneType' is not iterable

Hass.io 0.75.3 running on ResinOS 2.3.0+rev1. Supervisor 125.

Handle processing of Amazon Switch Account page for login process

For at least one German user attempting to login, the login process brings them to the switch account page. This page apparently relies on Javascript to render the switch user box and log off option, which means bs4 can't read the form option to process and our login script can't find a <form> to pass data to.

Based on reviewing the captured html page, this can be partially replicated by disabling javascript and going to the appropriate switch account link from the main Amazon page which demonstrates a broken page without selectable info.

I think the region is a coincidence and likely could happen to any account.

The equivalent US switch account url is here

We need to explore options (best to worst in my mind)

Simulate click to select HA configuration account.
Forcing logout and manually re-signing in (I suspect this issue only occurs because of one successful login, but I'm not sure what would trigger this after the login so this a temp solution)
Manual workaround (perhaps logging out all accounts option in Amazon.com, assuming it exists).

EDIT: 1 may not be possible after some further looking; the post request is being protected by a data and crsf token which is generated by the JavaScript. 2 will probably work as it's a simple get to https://alexa.amazon.com/logout which then 302s to the Alexa login page.

Device Owner problem with multiple Echos in household

Background: I have 4 Amazon Echos in the house. 2 of them were originally activated by me, 2 of them were activated by my wife. Once activate, you can attach multiple accounts to a device. All 4 of the devices are shown in my Amazon account, and my wife's account. When we are at a given device we can say "Alexa, switch accounts". This way my wife can play her music and I can play mine.

Problem: This script will successfully pull all 4 devices and show them, however, I can't control the ones originally activated by my wife. The reason is because this script is making the web request with my wife's deviceOwnerCustomerId (reported by Amazon as the owner), but I'm signed in as myself (a different deviceOwnerCustomerId). If I change the code to always use the deviceOwnerCustomerId of myself (hardcode) the script works fine.

Solution: I believe the fix for this is this component should access https://alexa.amazon.com/api/bootstrap and pull the deviceOwnerCustomerId of the current user. It would only need to do this one time (perhaps after it stores the cookies). It should then use this deviceOwnerCustomerId to make the calls, and not try to make the calls for another user that is not signed in. At the very least, perhaps a config option can be setup so that you can override the deviceOwnerCustomerId for all the calls.

This is not the same thing as #37 where they are asking for multiple logins.

I understand this code was copied from another library - I actually contacted the author of that plugin as well.

Alexa media_player component stays idle when playing playlists from spotify

Hi,

When I run a spotify playlist in my HA automation for example, the media_player.echo_plus_bathroom stays in state standby in HA even though music is correctly playing. If I say alexa play chillout lounge on spotify the same happens, no feedback on HA.

- service: media_player.play_media
   data:
        entity_id: media_player.echo_plus_bathroom
        media_content_id: Chillout Lounge
        media_content_type: SPOTIFY

It only happens for playlists apparently, and the Alexa App on my phone doesn't get feedback either. I'm running this in France

Alexa groups are not working

When sending an command to a single echo this works,
but sending an command to a alexa group of devices this won't work.

This problem concerns tunein and spotify.
Not sure if other services are affected.

Call service code:

{
  "entity_id": "media_player.uberall",
  "media_content_id": "Release Radar",
  "media_content_type": "SPOTIFY"
}

{
  "entity_id": "media_player.uberall",
  "media_content_id": "eins live",
  "media_content_type": "TUNEIN"
}

Possibility to control volume while media_player in standby mode

Thank you for your work put in to this component.

Please add possibility to change Echo device volume while it is in standby state.

Currently volume can be changed when media_player in Home Assistant is in state playing and paused but in order to make alexa_tts service constantly speak text at certain volume.

Volume, Tunein

Please add possibility to change Echo device volume and select tunein station via json?

Device does not show when playing radio stations

When a specific piece of media is being played it shows all the info as expected.

However if a Radio station is being played on the device the media information is just set to 'standby'. Can the radio information be exposed in the media information attributes?

Write Unit Tests for Alexa API

Create a test suite for the Alexa API to check for regression

Passing commands to Alexa

Some ppl here complained about Alexa handling their TTS requests as speech command like "Alexa switch light x on". I'd like to actively use this in my case to run my custom skill doing stuff. Background is I want Alexa to say something when I need it to, but more as question where I can reply to and the reply triggers something. Example is I enter the home zone and get a "Welcome home, do you want your PC being switched on?" "Yeah/Nope" and it passes my answer to Home Assistant doin specific stuff. TTS does not allow that but passing my skill a secific command I can realize such behavior. Can this "mis"-behavior be added as a feature maybe?

Move the Alexa API to a standalone pip module.

JSONDecodeError: Expecting value: line 8 column 1 (char 7)

Apologies in advance, I haven't fully fixed this yet, but wanted to report the issue and give you my findings.

As documented in the community thread, the Alexa component will fail to load with a JSONDecodeError.

The cause of the error is in AlexaAPI.get_devices because a valid JSON is not being returned from the Alexa device page, but instead an Amazon sign-on page is being returned.

This is because the login method has an incorrect test.

            if 'devices' in post_resp.text

This fails because the sign in page does contain 'devices' so will give an incorrect true condition. The test should instead check for a json, which apparently in python requires a try clause.
(edit fixed a bug in the code)

            try:
                    from json.decoder import JSONDecodeError
            except ImportError:
                    JSONDecodeError = ValueError
            try:
                post_resp.json()['devices']
            except JSONDecodeError:
                status['login_failed'] = True
            
            if ('login_failed' in status and status['login_failed']):
                _LOGGER.debug("Log in failure.")
            else:
                status['login_failed'] = False
                status['login_successful'] = True
                _LOGGER.debug("Succesfully logged in.")

This should fix the login test, but it doesn't fix it completely as the AlexaAPI.get_devices call may still get another sign-in page. (This is where I had to stop; not sure why the existing login session isn't enough). I think the solution would be to have the get_devices page (and probably all API calls) automatically login if it detects it's getting the sign-in page.

I'll probably try to implement it and will submit a PR if I get to it before you.

TypeError: 'NoneType' object is not subscriptable

I managed to get as far as successfully configuring the component but then it gets the below error which repeats approximately every 50 seconds.

018-08-08 18:27:45 DEBUG (SyncWorker_14) [custom_components.media_player.alexa] Using cookies to log in.
2018-08-08 18:27:47 DEBUG (SyncWorker_14) [custom_components.media_player.alexa] Logged in.
2018-08-08 18:27:47 DEBUG (SyncWorker_14) [custom_components.media_player.alexa] Log in successful with cookies
2018-08-08 18:27:47 DEBUG (SyncWorker_14) [custom_components.media_player.alexa] Setting up Alexa devices
2018-08-08 18:27:54 ERROR (MainThread) [homeassistant.core] Error doing job: Future exception was never retrieved
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/concurrent/futures/thread.py", line 56, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/config/custom_components/media_player/alexa.py", line 192, in setup_alexa
    update_devices()
  File "/usr/local/lib/python3.6/site-packages/homeassistant/util/__init__.py", line 325, in wrapper
    result = method(*args, **kwargs)
  File "/usr/local/lib/python3.6/site-packages/homeassistant/util/__init__.py", line 325, in wrapper
    result = method(*args, **kwargs)
  File "/config/custom_components/media_player/alexa.py", line 164, in update_devices
    update_devices, url)
  File "/config/custom_components/media_player/alexa.py", line 228, in __init__
    self.refresh(device)
  File "/config/custom_components/media_player/alexa.py", line 267, in refresh
    self._media_is_muted = self._session['volume']['muted']
TypeError: 'NoneType' object is not subscriptable

If there is any further info you require let me know.

Regards,
Michal

Reduce polling across component

The get_last_called Alexa request is not cached and will be pulled by every media_player refresh. This can result in excessive load if there are multiple echos on an account.

No platform alexa in HA with new release 1.0.0

Hi,

I downloaded the new version 1.0.0 and copied the folder alexa_media under custom_components and restart hassio.
afterwards I have the following error in log and no media_player components. What is to do? I missing a installation instruction in Readme.md or Wiki.

Queuing volume and mute in background

Also currently I have a flow in node red that sets the volume on all media devices to 10% at night and 30% during the morning. It would be great if I could set this and have it queued in the background. As well as possibly queueing mute state?

Originally posted by @niemyjski in #57 (comment)

Run through HA linter

TTS is not working with new version 1.0.0

Since update to version 1.0.0 on hassio 0.88.1 alex_tts seams to be not working ... with last version before 1.0.0 all was fine ... any hint or advice?

Feature Request: Device filter

Currently I have your 0.9.0 script working, mainly for TTS push messages on NAS device errors. Problem I have now is that my amazon account is polluted with devices I stopped using ages ago like a 1st gen Fire TV Stick and "Mobile Devies" and a "This device". A way to stop enlisting these would be very nice.

Cannot use if/else statement in message

I was trying to send a message to my Echos using if/else statement to change message depending on certain factors - like you can using the notify services. I assumed it would work with your custom alexa media player but it does not. Not sure if it can be worked in or not.

Example:

media_player.alexa_tts
{
"entity_id":"echo"
"message": "{% if ('sensor.somesensor', true) %}this{% else %}that{% endif %}"
}

[Feature Request] Support for multi room groups

Thank you for your great component
It would be great if the alexa.py component could support alexa multiroom groups.

NoneType errors where component does not generate and capture malformed json

Sep 01 16:58:42 raspberrypi hass[11271]:   File "/home/homeassistant/.homeassistant/custom_components/media_player/alexa.py", line 203, in update_devices
Sep 01 16:58:42 raspberrypi hass[11271]:     bluetooth = AlexaAPI.get_bluetooth(url, login_obj._session).json()
Sep 01 16:58:42 raspberrypi hass[11271]: AttributeError: 'NoneType' object has no attribute 'json'

Should be a simple fix to move the .json into the API like get_devices.

EDIT: Added other similar json errors.

Sep 01 10:20:56 raspberrypi hass[11271]:     alexa_clients[device['serialNumber']].refresh(device)
Sep 01 10:20:56 raspberrypi hass[11271]:   File "/home/homeassistant/.homeassistant/custom_components/media_player/alexa.py", line 321, in refresh
Sep 01 10:20:56 raspberrypi hass[11271]:     session = self.alexa_api.get_state().json()
Sep 01 10:20:56 raspberrypi hass[11271]: AttributeError: 'NoneType' object has no attribute 'json'

Error introduced in v0.10.0

I've been using this custom component very successfully for several months—Alexa TTS is so awesome.

However, updating to v0.10.0 seems to have broken the component for me and HASS throws the following error:

2019-01-25 16:13:41 ERROR (MainThread) [homeassistant.core] Error doing job: Future exception was never retrieved
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/concurrent/futures/thread.py", line 56, in run
    result = self.fn(*self.args, **self.kwargs)
  File "/config/custom_components/media_player/alexa.py", line 201, in <lambda>
    track_utc_time_change(hass, lambda now: update_devices(), second=30)
  File "/usr/local/lib/python3.6/site-packages/homeassistant/util/__init__.py", line 315, in wrapper
    result = method(*args, **kwargs)
  File "/usr/local/lib/python3.6/site-packages/homeassistant/util/__init__.py", line 315, in wrapper
    result = method(*args, **kwargs)
  File "/config/custom_components/media_player/alexa.py", line 235, in update_devices
    update_devices, url, authentication)
  File "/config/custom_components/media_player/alexa.py", line 318, in __init__
    self.refresh(device)
  File "/config/custom_components/media_player/alexa.py", line 358, in refresh
    self._last_called = self._get_last_called()
  File "/config/custom_components/media_player/alexa.py", line 448, in _get_last_called
    if self._device_serial_number == self.alexa_api.get_last_device_serial():
  File "/config/custom_components/media_player/alexa.py", line 1008, in get_last_device_serial
    if last_activity['activityStatus'](0) != 'DISCARDED_NON_DEVICE_DIRECTED_INTENT':
TypeError: 'str' object is not callable

alandtse / alexa_media_player Goto Github PK

alexa_media_player's People

Contributors

Stargazers

Watchers

Forkers

alexa_media_player's Issues

last_alexa() from alexa_remote_control.sh

Recommend Projects

Recommend Topics

Recommend Org

`last_alexa()` from alexa_remote_control.sh